Though the term “Solid State lonics” was named originally by the present author early in the 1960s, solid ionic conductors which are involved in the field of Solid State lonics have been recognized since the mid-nineteenth century. However, room temperature high coniuctivity solid ionic conductors have been studied only after the 1960s. A field of Solid State lonics covers the stuaies of all phenomena of ions in solids, especially solids exhibiting high ionic and electronic-ionic mixed conductivities at fairly low temperatures below their melting points. The applications of these solids are also included in this fiela. Here, only the early history of high conductivity solid ionic conductors which have the longest history in the field of Solid State Ionics is described briefly.

Solvent-free polymer electrolytes and polyelectrolytes are usually studied at quite high ionic concentrations, (into the range above 1M). Under these conditions, correlation effects arising from ion-polymer and ion-ion interactions are expected to be important in the mechanism of conductivity. We sketch some specific ionic effects, separating those actittg on the mobility from those effecting carrier concentration. Mobility effects include reduction of the fluidity due to the effective cross-linking by cations, screening of applied fields due to high ionic concentrations, frictional drag due to counterion motion, and in some polymer hosts, lowered local availability of cation solvation sites due to reduction of the number of coordinating basic oxygens. Reduction of the carrier density from its stoichipmetric value can be discussed in terms of a generalized ion-pairing model. Though the concentrations usually studied are so high that Debye-Huckel theory is invalid and the stoichiometric average cation-anion separation is smaller than the Bjerrum length (a situation in which ordinary electrolyte theory considers all ions paired), nevertheless consideration in terms of contact ion pairs, solvent separated ion pairs and mean stoichiometric separation can be used to compute the effective concentration of carriers. Estimates based on an electrostatic continuum, cavity model for the binding energy of a pair describe the reduction of effective carrier number observed in poly (propylene oxide) materials.

Structural properties, single-particle dynamics, and the charge transport are studied in superionic conductor Ag2Se using the molecular dynamics (MD) technique. The calculations are based on a model of interionic potentials in which ions interact through Coulomb interaction, steric repulsion and charge-dipole interaction due to the large electronic polarizability of the selenium ions. Structural and dynamics correlations are studied at five temperatures in the superionic phase. Among the structural correlations the results are presented for partial pair correlation function, coordination numbers, bond angle distributions and wave-vector dependence of the Bragg intensities. Detailed comparison with neutron and x-ray single crystal diffraction experiments. The calculated temperature dependence of the self-diffusion constant of silver is in good agreement with the tracer diffusion measurements. The spectra of velocity autocorrelation functions and the frequency dependent ionic conductivity are calculated. The Haven's ratio is also in good agreement with experiments.

Effective interatomic potentials consisting of two-body (steric effect, charge transfer and charge-dipole interactions) and three-body covalent forces are proposed for GeSe2. Using these interaction potentials in MD simulations, the nature of short-range and medium-range order is investigated in glassy and molten GeSe2. All the features in the static structure factor, S(q), including the first sharp diffraction peak (FSDP), are in good agreement with experiments. The FSDP arises from Ge-Ge and Ge-Se correlations between 4-8Å, and the anomalous decrease in its height on cooling is due to frustration enhanced by the increased density.

In solid ionic conductors, the hopping motion of the mobile charged defects is strongly influenced by their mutual repulsive interaction. Therefore, their jump diffusion ist not a random process. Instead, correlated forward-backward hopping sequences are a common phenomenon, constituting the elementary steps of jump relaxation. The microscopic dynamics of the relaxation is described in a simple model which yields the frequency spectrum of the hopping motion. With the help of this function, it is possible to explain experimental spectra displaying the so-called “universal” dynamic response. In particular we reproduce the frequency dependent ionic conductivity of fast ion conductors and the arcs in their complex-conductivity and complex-permittivity planes. The model also explains the non-BPP-type behavior of spin-lattice relaxation times and the broad components of quasielastic neutron scattering spectra. Examples are presented and discussed.

A model based on ion hopping in potential double-wells is proposed to explain the non-Debye dielectric response in solids. Relying on some assumptions, an attempt is made to remove the “average” nature of previous diffusion theories. This results in a distribution of activation energies, G(E), which decays exponentially on both sides of some given value E0. It is shown that (a) the existence of a dielectric loss peak is a result of the decay of G(E) for E > E0, (b) the constant-phase-angle behavior above the loss peak is associated with the decay of G(E) for E < E0, and (c) G(E) can produce all the main features of the empirical Havriliak-Negami function. An interesting property of this G(E) is that it broadens with increasing temperature, consistent with many experimental observations.

Li+ ions are highly mobile in LiCl-Li2O·2B23 glasses to the extent that the decou-pling of conductive modes of motion from the viscous modes, assessed at the glass transition temperature Tg by a relaxation time ratio, is one of the greatest known. The interpretation of the structure of this glass system and its relation to conductivity decoupling is not very advanced at this time but there seems to be a distinction drawn between the chloroborate structure and the structure of the corresponding AgCl-Ag borate and AgI-borate glasses. The latter have frequently been discussed in terms of α-AgI percolating clusters although the available thermodynamic evidence indicates chemical ordering (negative deviations from ideal mixing) as opposed to any sort of clustering (positive deviations from ideal mixing). The LiCl-Li2O-B2O3 system is interesting to us because simple transferable effective pair potentials are available for all species in the system and ion dynamics computer simulations should be capable of giving useful insights into the structure, energetics, and dynamics of the system. We present diffusivity data as a function of temperature for several compositions in the LiClLi2O·2B2O3 system and observe that the Li+ mobility remains high below the simulated, glass transition. Surprisingly, the Cl− anion mobility also remains high, raising a question about anion transport numbers in these glasses. Computed conductivities agree with laboratory data in the liquid state but exceed laboratory data in the glassy state in the direction expected from the high fictive temperature of simulated glass. Deviations from additivity in mixing energy in this system show a weak tendency to clustering of LiCl in the structure which suggests a need for laboratory mixing enthalpy studies.

This paper considers the results of NMR relaxation studies of fast ion glasses in the light of variable frequency electrical conductivity measurements. We show that the low activation energy observed for relaxation processes studied at constant frequency below the temperature of the T1minimum is reproduced by the corresponding constant frequency conductivity measurements. Other characteristics of the constant frequency conductivity in this low activation energy regime, such as the unphysically low pre-exponent, match with the corresponding observations for NMR measurements. Since the activation energy observed at constant frequency in the frequency-dependent regime for conductivity depends on the characteristic departure from exponential relaxation for the conductivity process, we conclude that the NMR activation energy is likewise a simple consequence of the nonexponential character of ionic relaxation in the glass. In this case, the low activation energy attributed to processes probed by NMR relaxation is simply a misinterpretation since it is found, in the wide frequency range studies available to admittance bridge measurements, that all elements of the relaxation spectrum have essentially equal activation energies.

Electron transport in mixed conductor redox electrolyte membranes: anion/electron charge carriers, and cation/electron charge carriers are analyzed by irreversible thermodynamic methods appropriate to electrolyte transport. The generalized force for electron transport is the gradient of the Fermi level of electrons in electrolyte phases. The generalized force for ion transport is the gradient of the ionic electrochemical potential. New results include current-time, current-voltage, small amplitude and DC bias impedance functions.

Starting from a Master Equation description of particle hopping between localized sites in a disordered system, we systematically analyze the influence of spatially-correlated hopping (memory) on the excitation transport. Correlations are introduced by accounting exactly for the hopping among a finite cluster of sites which are coupled in an uncorrelated manner to the remaining sites outside the cluster (background). Our results for quantities such as Ps(t). the probability of being at site s at time t for given initial conditions, compare well with simulation results on systems with both e−r and r−6 hopping rates, the agreement improving systematically with increasing cluster size. A simple interpretation of the time development of the cluster-background coupling is seen to be analogous to the overlapping-sphere construction of the percolation path.

A comprehensive time independent theory of the solid electrolyte(SE)–metal electrode interface is presented, using the assumption that cations are the only mobile charge carrier in the electrolyte. The temperature and the dc current across the SE are the only free parameters for the solutions along with three intrinsic parameters which depend on the properties of the particular system. The phenomenological model of the interface double layer is based on the Gouy–Chapman–Stern model.

The numerical solutions of the theory for porous tungstenzeolite system enables us to predict most of the properties in the SE system such as;the current–overpotential characteristics, the capacitances of the double layer, concentration profile in the diffusion layer, the potential profile across the interface, electrochemical exchange current, and the potential of zero charge(PZC) etc. An experimental technique to measure the PZC of SE systems has been also proposed.

The thermodynamic point defect concentrations for both ionic (majority) as well as electronic (minority) charge carriers are considered for different types of heterogeneities appearing in ionically conductive solid materials systems. Particularly, the electrical conductivity is discussed for different types of composite electrolytes. Quantitative space charge arguments are shown to be able to explain a variety of different phenomena such as: unusual enhancement of ionic conductivity in two phase samples and in polycrystalline materials, additional conductivity effects in microsized systems; change from cationic to anionic conduction due to heterogeneous doping; simultaneous enhancement of n- and p-conductivity in different electrolyte/ alumina samples, catalytic effects in composite electrolytes as well as chemical interface effects.

For some superionic conductive membranes the speed of diffusion of ions is enough to produce a measurable electric field in a magnetic field, which is situated perpendicularly to the di rection of diffusion. This phenomenon may be used in the manufac turing of different kinds of sensors.

The superionic conductors lithium nitride and lithium phosphide and the semi-metallic conductor lithium arsenide were synthesized through elemental reactions of lithium melt with nitrogen gas and phosphorus or arsenic powders. A high mobility of the lithium ion was found in this class of compounds. These compounds are hard and brittle and have a dark brown color. The structures of these compounds were studied using x-ray diffraction techniques. All three compounds crystallize in a hexagonal structure with a P6/mm space group. The lattice dimensions expand anisotropically as the size of the anion increases.

The ionic and electronic conductivities of the compounds were studied using reversible lithium electrodes and an ion blocking molybdenum electrode. The temperature dependent conductivity measurements (Arrhenius plot) show a high ionic and a negligible electronic conductivity for both lithium nitride and lithium phosphide. However, lithium arsenide shows semi-metallic behavior with a high electronic conductivity. A higher degree of covalency in the ct direction and a more ionic character in x-y plane were observed. The anisotropic nature of the chemical bonds in these compounds provides a shallow potential well in x-y plane and allows the correlated motion of lithium ions.

Lithium phosphide and nitride are superionic conductors and useful materials for application in batteries and sensors. Lithium nitride decomposes below 0.5 volts and can be used in low voltage and zero current devices. Lithium phosphide is stable up to 2.2 volts and can be used for higher voltage devices. These compounds decompose in contact with water and are air sensitive.

Oxide gels are actually hydrous oxides. Water adsorption and dissociation occur at the surface of the oxide network leading to charged particles surrounded by an acid or basic aqueous medium. They can be considered as particle hydrates. Fast proton conduction is observed arising from proton diffusion through the adsorbed water layers. Proton conductivity is actually strongly related to the water adsorption isotherm.

Ion exchange readily occurs at the oxide-solution interface when the gel is dipped inside an aqueous solution. Some inorganic gels exhibit a layered structure similar to that of sheet silicates. Ion exchange can then be described as an intercalation process. It can be used as a synthetic route toward new compounds.

Electrochemical insertion within the oxide network is highly reversible. This is due to the open structure of the gel and the mixed valence behavior of the transition metal oxide. Reversible cathodes or electrochromic layers have been made from V2 O5, W03or TiO2gels. Thin films or pressed pellets can be easily made from oxide gels. They are therefore good candidates for making "all gel" micro-ionic devices.

Increased interest in ion conducting thin films and thin-film devices is evidenced by the number of papers on this topic presented at the recent Garmisch conference [1[. Magnetically enhanced (magnetron) sputtering [2,3] is an important technique for depositing these as well as other kinds of ceramic thin films. Two advantages of magnetron over conventional diode sputtering are the increased sputtering rate, which is important in the deposition of ceramic materials, and the reduction in electron bombardment of the substrate and growing film. In this paper, we describe our high-vacuum film growth chamber and present the results of tests of this system in fabricating films of lithium ion conducting glass electrolytes, TiS2, and vanadium oxide by reactive dc and rf magnetron sputtering. The results of combining these films into a thin film lithium battery also are discussed.

The pyrochlore solid solution Gd2(Zrx Til−x)2O7, was found to be an attractive system for investigating the relationship between composition, structural disorder and ionic conductivity. Both cation and anion order parameters were found to decrease systematically with increasing substitution of Zr for Ti leading ultimately to intrinsic fast oxygen ion conductivity in the solid solution. The degree of intrinsic disorder was determined quantitatively from doping experiments and was found to be equal to l.0×lO39 exp(-O.24±0.03eV/kT)cm−6sfor x = 0.3 and substantially larger for higher values of x. Oxygen vacancy mobilities, on the other hand, were found to be relatively independent of x with values of μv, = 0.15exp(-0.78 ± 0.02 eV/kT)cm2V−1s−1. These, and more recent results, on Y2 (ZrxTil−x)2O7, are discussed in the context of the similarities between the pyrochlore and fluorite phases.

Hydrogen has been found to exhibit high mobility in several classes of solids, which also have appreciable electronic conduction.

The structure and properties of such mixed -conductor materials will be dicussed. In addition, several types of applications, based upon their special properties, will be described.

A number of composite polyphase electrode systems employing the mixed-conducting matrix principle have been proposed in this laboratory as alternatives to single lithium alloy systems for use as negative electrodes in both room and intermediate temperature lithium-based cells. Further, a composite hydroxide ion conducting electrolyte has been considered for use at intermediate temperatures which utilises liquid lithium hydroxide formed in situ and contained within a lithium aluminium oxide ceramic matrix.

Both these, and other, composite electrochemical cell component systems serve to overcome certain disadvantages found in simpler, non-composite designs. In order to operate effectively, the electrode and electrolyte systems described here must satisfy certain thermodynamic, kinetic and mechanical criteria. In this work, these criteria are discussed, using, among others, the examples given above, with particular emphasis on the thermodynamic phase stability conditions which must be satisfied by a viable composite electrode or electrolyte system.

The effects of halide substitution on ionic conductivity in potassium borate glasses are different from those observed in their Li and Na analogs. Physical property and Raman spectroscopy results suggest that this behavior is associated with differences in the glass network structure induced by the presence of different interstitial alkali cations.

Raman measurements on the Gd2(ZrxTi1−x)2O7, solid solution show increasing peak widths as the composition is varied from Gd2 Ti2O7, to Gd2 Zr2,O7. The increase in peak width is interpreted as reflecting an increase in structural disorder with increasing x in this pyrochlore structure material. Structural disorder occurs readily in such oxides by the formation of cation anti-site defects and anion frenkel defects. We have previously observed that this leads to a marked increase in oxygen ion conduction with increasing x. Preliminary results also indicate that certain peaks in the Raman spectra are characteristic of the end members of the solid solution and may occur at the same frequency independent of composition. The Raman results are discussed in relation to earlier x-ray diffraction and ionic conductivity data obtained in this group.